In offline validation, combined relevance-based secondary-AOI selection matched children's next attended AOI more often than random or single-distance selection¶
Claim · All claims
Evidence · 1 study · causalr2·q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q2 i? The combined relevance approach (0.2 semantic + 0.4 spatial + 0.4 temporal distance) achieved the highest next-AOI prediction accuracy under both proxy labels (0.209 expert-annotated, 0.215 future-gaze), above random and each single-distance method. → Zekun Wu 2026
Evidence¶
Zekun Wu 2026¶
Zekun Wu, Man Su, Huiyong Li, Tomohiro Nagashima, and Anna Maria Feit. (2026). Gaze-Informed Proactive AI Assistance for Children Exploring Picture Books. https://arxiv.org/abs/2607.00445 doi:10.48550/arxiv.2607.00445
q2 · i? · causal · r2
Offline validation predicting the next AOI on children's own exploration paths, with 12 to 15 candidate AOIs per image; the article notes absolute accuracy is expected to remain low because next looks are only partly determined. Printed accuracies are proportions, not effect sizes.
"As indicated in Table 2, the combined relevance approach achieved the highest accuracy under both proxy-labeling approaches."
Discussion¶
Related Claims¶
- In offline validation, the HMM timing model detected children's attention moments more accurately than fixed-interval and dwell-time triggers — related
- Attention allocated to the secondary AOI remained limited and did not differ significantly between conditions in duration or time to first fixation — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
combined-relevance-selection-aligns-with-next-attention |
title |
In offline validation, combined relevance-based secondary-AOI selection matched children's next attended AOI more often than random or single-distance selection |
description |
In offline validation, combined relevance-based secondary-AOI selection matched children's next attended AOI more often than random or single-distance selection |
status |
review |
evidence_strength |
weak |
newest_evidence |
2026 |
recheck_by |
2029 |
generated |
by process:wiki-ingest on 2026-10-10 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — zekun-wu-2026 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.